Li
Ye
,
Constantine
Shuniak
,
Razanne
Oueini
,
Jenay
Robert
and
Scott
Lewis
*
Department of Chemistry, University of South Florida, Tampa, Florida 33620, USA. E-mail: slewis@usf.edu
First published on 23rd May 2016
A well-established literature base identifies a portion of students enrolled in post-secondary General Chemistry as at-risk of failing the course based on incoming metrics. Learning about the experiences and factors that lead to this higher failure rate is essential toward improving retention in this course. This study examines the relationship between study habits and academic performance for at-risk students in General Chemistry. Students who were in the bottom quartile of SAT math scores were identified as at-risk students. The study habits of General Chemistry students, both those identified as at-risk and those not identified were measured by text message inquiries. The text message asked ‘‘Have you studied for General Chemistry I in the past 48 hours? If so, how did you study?” twice a week throughout a semester. Student responses to the messages were used to calculate the frequency of studying throughout the term. The results from a multiple regression analysis showed that high frequency of studying could mitigate the difference between at-risk and non-at-risk students on final exam scores. Additionally, the quality of studying for six at-risk students was analyzed by student interviews in concert with their text message responses. The results indicated that the quality of studying is not necessarily linked to frequency of studying and both quality and frequency can play a role in at-risk students' academic performance. The results presented offer a path for at-risk students to succeed in General Chemistry and the methodology presented offers a potential avenue for evaluating future efforts to improve student success.
Among those predictors, there is a long history of using SAT math score to predict students' academic performance in chemistry courses in the literature. SAT math is a component of the SAT, a standardized college-entrance test commonly administered in secondary school. SAT math is designed to measure quantitative reasoning including problem solving, modeling and algebraic structure (College Board, 2016). SAT math was strongly associated with student academic performance in chemistry; students who have low SAT math scores are more likely to have low academic performance in chemistry courses (e.g., Pickering, 1975; Spencer, 1996; Lewis and Lewis, 2007). The cut-off scores of SAT math used to determine at-risk students in chemistry were varied due to the diverse incoming abilities of students among universities. Lewis and Lewis (2007) examined a range of SAT math cut-offs and found that the bottom 25% to 35% of the sample by SAT math made approximately 65% to 75% correct predictions in describing a student as at-risk. Combined the research to date indicates that students who enter General Chemistry with low SAT math are disproportionately likely to not succeed in the course.
In chemistry, the main practices that help improve at-risk students' success in chemistry have been reported as offering external remedial coursework (Meckstroth, 1974; Pickering 1975; Walmsley 1977; Bentley and Gellene 2005; Heredia et al., 2012), group activities (Mason and Verdel, 2001) or training programs for at-risk students (Shields et al., 2012; Hall et al., 2014). Remedial courses typically offer at-risk students lectures on preparatory chemistry content concurrently or consecutively with the regular lectures. Pickering (1975) reported providing a supplementary course for students who had SAT math scores of 610 or lower. This course taught students the solutions of diverse types of problems associated with the content students learned in the parallel chemistry lecture course. Results showed that students who attended the supplementary course had mean grades that were 0.29 (on a 4-point scale) higher than the comparable students who did not attend, and the difference was significant. Interestingly, Bentley and Gellene's (2005) study suggested that the effect of a remedial course depended upon students' SAT math scores. They offered multiple sections of an Introductory Chemistry course that aimed to teach vocabulary, concepts and problem solving skills for students scoring below 50% on chemistry placement test (CPT). Results showed that students with low scores on the CPT who took the Introductory Chemistry course finished with a grade in General Chemistry that was ¼ to ½ of a letter grade higher than their counterparts who did not take Introductory Chemistry. However this effect was only found for students with SAT math scores from 460 to 600, little or no effect was found for students below or above this range. The above studies focus on providing more repetition of course content for at-risk students. Similar to the general results suggested by the meta-analysis study, these articles report mixed effectiveness of these practices for at-risk students in chemistry. Even though positive effectiveness has been reported, since the above studies all used student grades as the outcome measures, the effectiveness could be partially attributable to how instructors assigned grades to students.
Other past work describes efforts to improve at-risk students' study skills through training programs or group activities. Hall and colleagues (2014) trained less prepared students who were in the bottom quartile of SAT math scores or lacked advanced placement (AP) courses in math and science by using a project called Science Advancement through Group Engagement (SAGE). SAGE was run concurrently with the regular lectures and implemented study group sessions focused on foundational chemistry knowledge with the aid of teaching assistants. SAGE also trained students with a self-regulated learning (SRL) approach. SRL encourages students to follow a study cycle of task analyses, planning, reflection and self-adjustment based on the value and meaning of their efforts. The results showed that the retention of SAGE participants was more than double that of the non-SAGE participants and historical group in chemistry sequence courses. By the fourth course, Organic Chemistry II, SAGE participants performed as well on final course grades as those students who had stronger high school backgrounds with more AP science and math courses and significantly higher SAT scores. Shields and colleagues (2012) implemented a transition program including extended-length recitations, peer-led team-learning (PLTL) study groups and peer-mentoring groups to help underprepared students who were in the bottom 25% of predicted scores based on ACT math, total of STEM AP test scores and online diagnostic scores. The study found the transition program helped the participants make significant gains in final general chemistry course scores that combined quiz scores, midterm and final exams in comparison to students who were in regular recitations only.
In both Hall et al. (2014) and Shields et al. (2012), at-risk students were trained with certain study skills or strategies and notable academic benefits were observed. However, neither of these investigations incorporated a measure of study habits so it is not possible to make a definitive claim that the interventions employed influenced student study habits. Another plausible explanation might be self-selection bias where participating students possessed higher motivation to succeed than the reference group from the onset of the study. Additionally, it has not been well established that the reason at-risk students struggle in General Chemistry is related to their study habits. That said, the notable benefits observed are cause for further investigation into the relationship between study habits and the academic success for at-risk students.
Sinapuelas and Stacy (2015) built on this model in an effort to characterize the study approaches of introductory, non-major college chemistry students. In this study, 61 students were interviewed at three time points throughout the semester. In the interview students were asked to describe the resources they used to prepare for exams and to elaborate on how they were used. The analysis of student responses led to the creation of the Learning Approaches for Chemistry framework that describes learning approaches in four hierarchical levels:
Level 1: gathering facts – students tend to memorize unrelated facts by scanning course materials, typically independently. Students do not monitor their own learning.
Level 2: learning procedures – students begin to make connections between pieces of information and try to work out practice problems. Students rely on others for answers, but they possess basic metacognitive skills such as assessing for procedural errors.
Level 3: confirming understanding – students evaluate and question data, form their own arguments, and work collaboratively with peers. Students assess their own knowledge based on their ability to justify and explain answers.
Level 4: applying ideas – students question data, try to use concepts to explain real-world phenomena, and act as “teachers” with their peers. Students possess advanced metacognitive skills such as assessing for gaps in conceptual understanding.
Levels 1 and 2 emphasize memorization, matching the description of surface level approaches; levels 3 and 4 emphasize content generation and application, matching the deep level approaches (Sinapuelas and Stacy 2015). Since this framework provides additional description to the surface-deep dichotomy and describes students' approaches while engaging in studying chemistry, this framework will be used to describe the quality of students' study habits in this work.
Specific to General Chemistry, Chan and Bauer (2016) divided students into high, medium and low affective groups using cluster analysis on the results of a survey measuring attitude, self-concept and motivation in chemistry. Surveys, open-ended questions and interviews were used to investigate students' study strategies used in the lecture and when preparing for exams. Students in the high group reported understanding the notes they took in the lecture more frequently than the low group, and the low group relied more on others for help when preparing for exams, analogous to the surface level learning description in Sinapuelas and Stacy’ article (2015). In addition, answers to the open-ended questions showed that the high group tended to be more confident about their study strategies while students in the low group felt less confident about their strategies and planned on changing their current study strategies, suggesting that confidence and studying strategies are related constructs.
Ye et al. (2015a) examined students' study habits of General Chemistry students outside the class via inquires sent through text messages. Students were characterized based on the types and frequencies of studying reported in their text message responses. Using cluster analysis, three patterns of studying emerged: students who knowingly do not study (Cluster 1), students who study in addition to the mandatory course components such as reading the textbook or practicing problems (Cluster 2) and students who primarily describe mandatory course components such as doing homework assignments as studying (Cluster 3). These three groups were compared on the measures of final exam and revised two-factor Study Process Questionnaire (rSPQ) (Biggs et al., 2001), an instrument used to measure students' study process with two sub-scales of deep and surface approaches. The results of ANOVAs showed that students in Cluster 2 were significantly higher on the final exam than the other two clusters. Students in Cluster 1 were significantly higher on the surface approach than the other two clusters, and Cluster 2 was significantly higher on the deep approach compared to Cluster 1. These results indicate that frequency of studying relates to academic performance in General Chemistry though the sample was not delineated for at-risk students. In reviewing the literature, no research exploring the role of at-risk students' study habits in the context of post-secondary chemistry was identified.
1. What is the relationship between study habits (frequency) and academic performance in college General Chemistry for at-risk students as compared to the larger remaining General Chemistry cohort?
2. What are the effective and non-effective study habits (frequency and quality) of at-risk students in college General Chemistry and what additional factors may explain the study habits employed?
The data collection spanned two semesters. First, in the spring semester, students were recruited from three of the four General Chemistry I classes on the first day of class. In the recruitment, the nature of the study was described to students. The participants would be asked to provide their cell phone numbers and would twice weekly receive a text message that asked the same question: “Have you studied for General Chemistry I in the past 48 hours? If so, how did you study?” The text messages would be sent at random times between 9 AM and 9 PM. Participants would be asked to reply to the message within 12 hours if possible. To encourage participation, students who replied to at least 80% of the text message inquires would be entered into a raffle for a $25 gift card at the end of the semester. The recruitment led to 301 students agreeing to participate in the study. The text message inquiry was sent out 28 times over the course of the semester. The text message responses from participants were collected and managed via a commercial online website. Student performance such as test scores, course grades, attendance and homework completion, along with demographics and SAT scores were collected from either university records or in-class records. Clickers were used for each class to record student attendance in the setting.
During the following fall semester, 28 at-risk students who replied to at least one quarter of the 28 text message inquires and were currently enrolled in General Chemistry II were invited via E-mail for a follow-up interview. Students who volunteered would be compensated with a $20 gift card. Six students volunteered and each was interviewed individually. The interviews covered three major themes: students backgrounds, e.g., major and prior chemistry coursework; elaboration on the study approaches reported through text messages, such as how the textbook was used; and questions that were related to students' approaches to learning, e.g., working with others, metacognition and affective factors. A complete list of interview questions can be found in the appendix. The interviews adopted a semi-structured approach. The lengths of the interviews ranged from 20 to 40 minutes. All data collection was carried out with the approval of the university's Institutional Review Board (IRB).
Variables | Non-at-risk | At-risk | Cohen's da |
---|---|---|---|
a Cohen's d = 0.2 (small), 0.5 (medium) and 0.8 (large). | |||
N | 384 | 153 | |
SAT math (mean ± SD) | 583 ± 52 | 476 ± 39 | 2.32 |
Test 1 (%) | 71.2 ± 14.0 | 63.7 ± 14.1 | 0.53 |
Test 2 (%) | 69.9 ± 15.9 | 63.1 ± 14.9 | 0.44 |
Test 3 (%) | 52.8 ± 15.2 | 41.2 ± 14.1 | 0.79 |
Final test (%) | 51.2 ± 15.2 | 44.8 ± 16.3 | 0.41 |
Attendance (%) | 82.3 ± 19.4 | 81.0 ± 20.3 | 0.07 |
HW (%) | 94.2 ± 14.9 | 95.8 ± 11.8 | 0.12 |
Course GPA | 2.78 ± 0.76 | 2.41 ± 0.75 | 0.49 |
To determine whether the group differences were significant, MANOVA analysis and univariate follow-up tests were conducted on the variables of each test, attendance, homework and course GPA listed in Table 1. Results of the MANOVA showed the group difference in means on the set of outcome variables was statistically significant with α = 0.05, F(7,529) = 10.243, p < 0.001, Λ = 0.881, which means the proportion of variance in the combination of outcome variables that was accounted for by the grouping variable was 12%. The size of the multivariate effect was estimated to be medium (ω2c = 0.10). The results of univariate follow-up tests revealed statistically significant group differences for each test and course GPA but not attendance and homework completion. Effect size measured by Cohen's d for comparisons on each individual variable are also listed in Table 1.
In sum, at-risk students performed worse on each single test and final course grade than non-at-risk student in the General Chemistry I course, but the effort measures such as attendance and homework completion were comparable. At-risk students were displaying as much effort as non-at-risk students but achieving less on the tests. These results support the method of identifying at-risk students based on SAT math as appropriate.
Course grades were converted into 4-point scale numbers for computing averages. For each student, percentage of attendance was calculated as the number of days the student recorded a clicker response divided by the maximum number of days the student could record a response in the semester. Homework completion in the course was measured using percentages of completion of homework. Instead of homework grades, percentages of the homework completion were used to measure student effort. In order to make the scales of the four outcome variables consistent, test scores were transferred into percentages.
To examine the relationship between study habits and academic performance in college General Chemistry for at-risk students as compared to the non-at-risk students, scatter plots showing relationship between study percent and final exam score for the two groups were constructed. To determine statistical significance, a multiple regression was conducted where SAT math score, study percent, and the interaction between study percent and SAT math score were used to predict students' final test scores. The reason for using final test scores in the regression model is that the study percentages represent studying across the entire term and the final exam was the only cumulative test.
The six interviews were transcribed verbatim using an open coding method. First, four chemical education researchers coded the transcripts independently; each person was assigned to code one to two distinct transcripts to describe all the themes present. The separate themes identified were compiled and the researchers discussed the similarities and differences among their themes to create a unified code list. Finally, two of the researchers coded the six transcripts independently based on the unified code list using NVivo 11.1.1 software. Upon completion of coding, disagreements between codes were discussed until consensus was reached.
Variables | Non-at-risk | At-risk | Cohen's da |
---|---|---|---|
a Cohen's d = 0.2 (small), 0.5 (medium) and 0.8 (large). | |||
N | 94 | 28 | |
SAT math (mean ± SD) | 586 ± 51 | 483 ± 28 | 2.50 |
Test 1 (%) | 69.9 ± 15.1 | 62.4 ± 13.4 | 0.53 |
Test 2 (%) | 72.0 ± 14.6 | 66.2 ± 11.2 | 0.45 |
Test 3 (%) | 53.2 ± 15.3 | 42.0 ± 11.9 | 0.82 |
Final test (%) | 49.1 ± 13.5 | 45.7 ± 15.8 | 0.23 |
Attendance (%) | 83.1 ± 17.0 | 81.6 ± 21.0 | 0.08 |
HW (%) | 94.0 ± 12.4 | 97.3 ± 6.2 | 0.34 |
Course GPA | 2.76 ± 0.72 | 2.50 ± 0.61 | 0.39 |
Study percent (%) | 46.6 ± 25.4 | 61.6 ± 29.3 | 0.55 |
First, noting that comparison between Tables 1 and 2 on the same variables shows that the selected sample and the broader population are very similar, supporting the ability of the sample to represent the population at least on the variables of interest. Second, the average study percent outside the class for at-risk students and non-at-risk students were 61.6% and 46.6%, respectively. It is interesting that the study percent for at-risk students was 15% higher than non-at-risk students, which means at-risk students reported studying more frequently outside the class than non-at-risk students in our setting.
For each student, the percent of text responses that used a particular study habit was calculated and the average for each study habit for the non-at-risk and at-risk students are presented in Table 3. For brevity, only the study habit codes that represent at least 5% of the text responses are shown. The data in Table 3 suggests that the study habits employed by the at-risk students did not differ from the non-at-risk students in terms of relative frequency; however, the at-risk students did employ each study habit at a higher rate.
Study habit | Non-at-risk students (%) | At-risk students (%) |
---|---|---|
Reviewed notes or PowerPoint | 18.3 | 21.8 |
Reviewed the textbook | 15.4 | 19.8 |
Online homework | 13.1 | 18.0 |
Practiced problems | 6.6 | 10.0 |
Previous exams or study guides | 5.8 | 9.0 |
Fig. 1 Scatter plots showing correlation between the study percent and final exam for non-at-risk and at-risk students. |
To further examine the differential relationship, a multiple regression model was run using SAT math, study percent, and the interaction between SAT math and study percent to predict final exam score. The interaction term was added to model the differential relationship. The multiple regression model is presented in Table 4 and suggests the linear best-fit equation of:
Final test = −0.339 + (0.00137 × SAT math) + (0.939 × Study percent) + (−0.00150 × SAT math × Study percent) |
Variables | b | Std. error | Beta | p-Values |
---|---|---|---|---|
Constant | −0.339 | 0.243 | 0.166 | |
SAT math | 1.37 × 10−3 | 4.28 × 10−4 | 0.617 | 0.002 |
Study percent | 0.939 | 0.383 | 1.798 | 0.016 |
Study × SAT math | −1.50 × 10−3 | 6.89 × 10−4 | −1.568 | 0.032 |
The prediction model statistically significantly predicts students' final test, F(3,118) = 5.39, p = 0.002, R2 = 0.121, with a medium effect size f2 = 0.14 (Cohen, 1988). All terms were significant (p < 0.05) except for the constant (Table 4).
The results indicate that both study percent and SAT math score are positively associated with final exam score. The interaction effect between study percent and SAT math is negative and significant, indicating that the differential relationship observed earlier is unlikely to be attributed to chance. The effect of study percent on final test score depends on students' incoming SAT math scores; in short, a high rate of studying can mitigate the impact of low incoming SAT math scores. To confirm that the results were also applicable for other students, the multiple regression model was also conducted on the participants who replied to the text messages at least once, with the same trend in results observed.
Fig. 2 shows a diagram plotted based on the regression equation. The lines represent the relationship between study percent and predicted final test scores when students have different SAT math scores using 50-point iterations in the range of 500 to 650 (representing the 15th to 93rd percentile in the sample). In general, higher SAT math score leads to a higher score on the final test. However, the differences caused by SAT math scores in performance on predicted final test scores for students change dramatically by frequency of studying for students. For at-risk students (math SAT < 515) the frequency of studying outside the class played a more important role in predicting final test scores than those with higher SAT math.
The differential relationship of frequency of studying with academic performance for different SAT math levels merits further study. One possible explanation is that students with different SAT math respond to earlier assessments in a different manner. For example, students with higher SAT math who perform well on the assessments throughout the term study at a relatively low frequency and continue to perform well on the final exam, possibly a result of seeing similar content in secondary school. However, when students of higher SAT math are not performing to their satisfaction, they respond by studying at a very high frequency. Students with lower SAT math may have an opposite relationship. Students with lower SAT math who perform well on early assessments may respond to the positive feedback by continuing to study at a high rate. However, if lower SAT students perform below their expectations, they may be discouraged and study less frequently as they do not expect to see a payoff from their efforts. This proposed explanation for the differential relationship essentially uses incoming SAT math scores and early academic performance as a proxy for students' self-efficacy. The role of self-efficacy in terms of study habits for at-risk students will be explored in the second research question.
(a) | ||||||
---|---|---|---|---|---|---|
Name | Jack | Ellie | Mary | Bella | Lucy | Ava |
a Numbers listed in the parentheses mean the number of times the study approach was mentioned in the text message responses. b “i” means the study approach only was mentioned in the interview. | ||||||
Gender | Male | Female | Female | Female | Female | Female |
Race | White | Asian | White | Hispanic | Hispanic | Black |
Major | Environ. Science | Physical Therapy | Pre-Medical Sciences | Biomedical Sciences | Biomedical Sciences | Biomedical Sciences |
SAT math | 510 | 460 | 430 | 510 | 490 | 510 |
Actual final test (%) | 29 | 26 | 64 | 58 | 38 | 46 |
Predicted final test (%) | 42 | 30 | 49 | 52 | 51 | 41 |
Final grade General Chemistry I | C | C | A− | B | C+ | B− |
(b) | ||
---|---|---|
Name | # Of text responses | Study approaches (frequencya) |
Jack | 21 | Homework (4), textbook (3), notes (2) and study in groups (ib) |
Ellie | 28 | Homework (4), practice tests (1) and study in groups (1) |
Mary | 24 | Notes (11), textbook (9), practice tests (6), study in groups (6), flash cards (3), practice problems (2), visit instructor (1) |
Bella | 25 | Homework (14), textbook (5), notes (2), practice tests (2), practice problems (1), study in groups (i) |
Lucy | 27 | Notes (18), homework (11), textbook (2), practice tests (1), online videos (1), study in groups (i) and visit instructor (i) |
Ava | 28 | Homework (4), notes (4), textbook (3), practice tests (1), online videos (1), study in groups (i) and flash cards (i) |
Ellie did all the homework assignments and tried to solve problems on the practice tests, but she spent a considerable amount of time stuck on problems:
“I go through the homework, I do the homework problems again… and then a test review I usually look at that and try to solve each problem... and then if I don't really get one problem, I sit there for like an hour, and I'm trying to like figure it out and I finally get it after five hours.” (Ellie)
Like Jack, Ellie also reported relying on friends in her studying: “In college, I don't really know anything on chemistry so I depend a lot on my peers” and “in General Chemistry, I had my friend like be there every step of the way and help me.” Ellie's feature of relying on others happened in different studying scenarios, for example, in the peer leading session:
“In the peer leading everyone is inputting their own ways of how to do it… but I like to get the right answer from the main person [peer leader], know I'm learning it right, rather than trying to figure it out and do it wrong and then I learn it wrong and remember it wrong.” (Ellie)
When she was not sure about a concept in her studying, she also still ended up seeking help from others: “I either like sat there and cried or I would go try to like find it online and see if they could explain it. And if I still didn't get it and I went to like one of my peers and I was like, ‘Hey, explain this to me’ and then after like a while I finally got it.” One anecdote in Ellie's interview further evidenced a surface approach to studying. When asked to nominate the most interesting things she learned in General Chemistry I, she responded: “The most interesting thing I learned… I don't know… it's bad but I don't really remember like what it was in General Chemistry, I learned.”
“I mean the concept of molality and molarity, I know this is kind of sad and embarrassing for me but it took me about a week to be able to distinguish the two. I'm a slow learner… I'm not confident because it'll take me forever to learn one topic.” (Ellie)
Interestingly, Ellie mentioned that she was more confident in the General Chemistry II course, a difference she attributes to having a different instructor. In terms of interest, both of them stated that were not interested in chemistry, “Like I'm a science fan. But chemistry, not so much” (Jack) and “not so much [interest] in chemistry.” (Ellie)
In summary, Jack and Ellie both describe surface approaches and low self-efficacy regarding chemistry, exhibited by the belief that they could not solve problems on their own, and a reliance on others as a coping strategy. In the Learning Approaches Framework for Chemistry, both Jack and Ellie provide indications of the first two levels by seeking to memorize facts and rely on others for answers. This could also explain the infrequent studying exhibited throughout the semester. As seen in Table 1, both Jack and Ellie finished with a C, the lowest possible grade available given the selection criteria of enrollment in the follow-on course.
“The homework, is kind of like my self-quiz… so I study a little bit and then do that [homework], so if I can do on my own, and tell myself I am doing good, and then would I need help, I have my notes for it, that is the concept I start reviewing.”
She marked things she didn't understand when studying and brought them to the instructor in the office hour. “I didn't go every week, but during test week, I make an appointment to go there at least once, sit with her, review things that I circle or mark that I don't understand.”
In the interview, Bella didn't articulate many details about how she used each study approach, but her answers to some of the questions projected that she adopted deep level approaches. For example, she said that when she was unclear about concepts, she searched online or went to ask peers, but she would not rely on them, in contrast to Jack and Ellie. “When I don't understand a problem obviously I go to my peers but I don't rely just on them and just study with them. Like I'll study maybe twice with them and then as the exam approaches closer I'll just focus on studying alone.” In addition, like Mary, Bella asked her instructor for help after making her own attempts. “I would go to her office hours during that week [the week of exams]. And then I would take my practice exam with me and then ask whatever I have problems with”.
Another theme in common for Mary and Bella is both of them liked to help others when studying in groups, and they thought helping others could help themselves learn better. When describing the interaction with others in a group, Mary likes tutoring others because it helps her too: “I prefer to study in group, …I end up knowing more so I tutor, I am like tutoring everyone, it helps me cause I teach it I know it.” Bella described a mutually beneficial relationship between her and her peers. “If you have a problem, you can go to your peers… They can explain it to you, and they have a problem, you can help them, also when you are helping people, you are kind of learning yourself.” In summary, both Mary and Bella provided a description that matches Level 3 where students develop their own understanding and use interactions with others primarily to confirm their understanding.
For the text messages, both Mary and Bella thought they were reminders of studying. Mary kept track of her text message responses and she used them to reflect on her study habits. “I found it [the text message project] was helpful… try to improve myself by writing things down, like what I would do differently, you really see what you are doing instead of just doing it.” Bella also expressed that text messages helped her to study more. “It [the text message project] did help me a lot and like doing a lot of stuff after class and reading the notes. So it did remind me to do all that kind of stuff.”
“I felt like it explained everything and then even in the end of the chapter it gave short block summaries of different concepts or whatever, so after you read them in depth, even if you didn't understand it in depth, you could flip to the back of the chapter and have like little sections like that, that helped a lot and especially I happened to be in a situation where I was cramming, I could always flip to those sections, so I really relied on the textbook.” (Ava)
Ava also described an active approach for lectures: “I print out the lectures ahead of time and I write my notes on the actual slides and then I can like actually point out what's important”. She described the interaction between herself and others as two types. First, in working with her peer leader, she shows self-efficacy in putting forth her own understanding: “I would ask my TA during the peer meeting, or if I didn't understand something I would go to him and be like… ok, well, if I do it this way, am I right or wrong, and that would help me.” Second, when working with her friends, she could help others and others could also help her. “So it's like I know A, you know B, and now we can put it together. In the situation now I did really well on the first test, so it's just like ok I can help somebody else”. In terms of approaches to learning, Ava shows signs of confirming understanding, analogous to Level 3 in the learning approaches framework, even with her infrequent reports of studying.
Third, when only one of the criteria is met, the quality of studying may matter more than the frequency of studying. Ava, who studied less frequently but adopted deep level study approaches performed better on her final course grade than Lucy, who studied more frequently but used surface level study approaches. Similarly, Ava over-performed her predicted final test score while Lucy under-performed her predicted score. According to the data analysis of the four groups in our study, we proposed a hypothetical model that both the frequency and quality of the study habits can be closely related to at-risk students' academic performance, and the quality of study habits might be more important than the frequency of the study habits. However, this model has to be tested by a bigger, more diverse sample in order to propose a generalizable claim. Determining the relative importance of quality versus frequency remains an open question that will be important to better understand how to assist at-risk students.
The interviews were consistent in that examples of good metacognitive skills and positive affective factors coincided with higher quality studying. For example, students' self-monitoring of their study approaches was only present for students who indicated deeper level studying approaches: Mary recognized that tutoring others helped her own learning and Ava realized that she needed to make connections among chemistry topics and not just memorize them as separate facts. Mary and Ava also kept track of their text message responses and utilized them to help promote their studying. In addition, high achieving students Mary and Bella showed more self-efficacy in learning chemistry and more interest in chemistry than low achievement students Jack and Ellie. Interestingly, both Lucy and Ava mentioned that they were not confident in General Chemistry I, but after General Chemistry I, the level of confidence increased because of studying. It seems likely that self-efficacy and frequency of studying are interrelated, where increases in one can beget increases in the other. In sum, students' metacognitive skills and affective factors such as self-efficacy and interest in chemistry can help us understand why and how at-risk students can succeed in chemistry. These traits are closely related to students' study habits (frequency and quality), which impacts their academic performance.
Additionally, the interviews conducted only examined factors related to student self-efficacy and metacognition in seeking to understand characteristics that can explain study habits. Other student factors such as time available to study, perceptions of knowledge generation or familial/social expectations for education may certainly prove to be relevant in understanding study habits. Finally, the interview cohort did not include students who failed or withdrew from first-semester General Chemistry and these students may provide unique, additional characteristics of study habits for at-risk students.
Furthermore, findings through text message responses and interviews suggested that both the frequency and quality of studying are important to academic performance for at-risk students. The results of this study lead to several implications for instructors who are teaching college chemistry courses. First, instructors should encourage at-risk students to believe they can succeed by studying (Cook et al., 2013). More importantly, instructional supports should be developed to promote at-risk students studying more and to develop deep level study approaches. In order to increase the frequency of studying, instructors can suggest students keep records of when and how they study in certain time ranges, and use those records to keep track of their study. The text message methodology presented here is one possible approach for doing so.
For the quality of studying, it is essential for instructors to help students develop deeper level study approaches. Instructors can provide specific guidance for at-risk students to use study materials, for example, by making annotated notes while reading the textbook or working the practice problems in the textbook. Likewise, for lecture notes, students can be encouraged to read the notes before the lectures and take their own notes during the lecture to support understanding the content instead of capturing all that is said. After the lectures, it is better to actively summarize or rewrite notes using a student's own words instead of only reviewing the notes taken. In terms of practicing problems, the Learning Approaches for Chemistry framework's emphasis on students generating their own understanding and using others primarily to confirm their understanding is prescient. Thus, efforts to promote students attempting to practice problems independently before comparing with an answer key or asking for help would be recommended. In terms of group work, the successful at-risk students presented here demonstrated independent learning by helping others in groups or using groups to confirm their understanding. The importance of explaining concepts when participating in a group matches learning theories and past research on how group work is effective (Webb 1989, 1992; Slavin, 1996). An instructional implication that follows would be the practice of assigning and rotating roles within the group, where one role has an explicit function of providing explanations when the group is called upon.
For researchers who are interested in designing interventions aimed at helping at-risk students, improving the frequency and quality of study habits are appropriate targets. Past research reviewed herein has described promising intervention techniques that may improve study habits. Future research can be aided by matching these interventions with measures to assess students' study habits with the methodology used here as one potential path for doing so. Another potentially fruitful area for research is to investigate the impact of pedagogy and classroom environment on students' study habits.
• What is your major? Why did you take General Chemistry I?
• What, if any, chemistry classes have you taken before General Chemistry I in college? In high school?
• Why did you take these classes?
• Are you satisfied with your performance in these classes?
• How would you describe your performance in previous chemistry classes?
• How would you characterize your study approaches?
• Please describe any changes in how you study when transitioning from high school to college.
• How important is studying with peers in high school versus studying with peers in college?
• How confident are you in chemistry?
• How satisfied are you with your resulting grade in General Chemistry I?
• If you could return to when you were in General Chemistry I, would you do anything different in your studying for General Chemistry I? If yes, what would you do?
• How is studying for chemistry different than studying for other classes?
2. Study habit text message clarification
• To what extent did participating in the text message project influence your study approach?
• If you did not respond to a text message, what was the reason?
• Describe how you used [X] in your studying. X = the approaches the students indicated in their responses. e.g. textbook, homework, peer activity…
• What study approaches did you think were helpful for General Chemistry I?
• What study approaches did you think were not helpful for General Chemistry I?
3. External study habit questions
• To what extent did the course instructor influence your study approach
• Was there a particular way the teacher presented the material that you liked a lot? Was there a particular way you did not like?
• How many peers in chemistry do you interact with? How important are these interactions? Describe the nature of these interactions, what types of discussions do you have with your peers in chemistry?
• What prevented you from studying for General Chemistry I?
• What factors were outside of the chemistry content?
• To what extent do you memorize content in General Chemistry I?
• Would you characterize the content in General Chemistry I as having one or a small set of themes or as a list of separate facts?
• How much of your studying for General Chemistry I was in practicing math examples versus conceptual understanding?
• What did you do when you were not sure about a concept in your studying?
• Do you think you had too many tests or not enough tests in General Chemistry I?
• Do you read about science/chemistry beyond what is covered in the course?
• What was the most interesting thing you learned in General Chemistry I? Why was it interesting?
• How did your study approach change between General Chemistry I and General Chemistry II?
• How confident are you in learning General Chemistry II now?
• How are you getting ready for your upcoming General Chemistry II test?
• Where do you like to study?
• Describe your ideal study environment.
This journal is © The Royal Society of Chemistry 2016 |